{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:NWMCIAQQX5TZ63VWJEHOD6CRRO","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"c37f5c14e9f42e31d8d269f3fac0c54cf428f24bcf8d0418e9e565b49210da13","cross_cats_sorted":["cs.AI","cs.CL","cs.LG","cs.SE"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CR","submitted_at":"2026-02-13T18:38:18Z","title_canon_sha256":"aac9a2b7a6cc1c7c0bb919def4b518c639cc06088b59ca415dd6b7dffbf9d1e0"},"schema_version":"1.0","source":{"id":"2602.13379","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2602.13379","created_at":"2026-06-12T01:08:22Z"},{"alias_kind":"arxiv_version","alias_value":"2602.13379v2","created_at":"2026-06-12T01:08:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.13379","created_at":"2026-06-12T01:08:22Z"},{"alias_kind":"pith_short_12","alias_value":"NWMCIAQQX5TZ","created_at":"2026-06-12T01:08:22Z"},{"alias_kind":"pith_short_16","alias_value":"NWMCIAQQX5TZ63VW","created_at":"2026-06-12T01:08:22Z"},{"alias_kind":"pith_short_8","alias_value":"NWMCIAQQ","created_at":"2026-06-12T01:08:22Z"}],"graph_snapshots":[{"event_id":"sha256:40ff6123d60e14118d99f29307cb071564ad69b4720b108d2b8b83314c2d24af","target":"graph","created_at":"2026-06-12T01:08:22Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2602.13379/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"LLM-based agents are becoming increasingly capable, yet their safety lags behind. This creates a gap between what agents can do and should do. This gap widens as agents engage in multi-turn interactions and employ diverse tools, introducing new risks overlooked by existing benchmarks. To systematically scale safety testing into multi-turn, tool-realistic settings, we propose a principled taxonomy that transforms single-turn harmful tasks into multi-turn attack sequences. Using this taxonomy, we construct MT-AgentRisk (Multi-Turn Agent Risk Benchmark), the first benchmark to evaluate multi-turn","authors_text":"Bo Li, Dawn Song, Minzhou Pan, Simon Yu, Weiyan Shi, Xue Lin, Xu Li, Yiyou Sun","cross_cats":["cs.AI","cs.CL","cs.LG","cs.SE"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CR","submitted_at":"2026-02-13T18:38:18Z","title":"Unsafer in Many Turns: Benchmarking and Defending Multi-Turn Safety Risks in Tool-Using Agents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.13379","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:ad1b0f8f088f32038e4b71eaa393f3c142b3d7343644daa2a384a9b5396e3430","target":"record","created_at":"2026-06-12T01:08:22Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"c37f5c14e9f42e31d8d269f3fac0c54cf428f24bcf8d0418e9e565b49210da13","cross_cats_sorted":["cs.AI","cs.CL","cs.LG","cs.SE"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CR","submitted_at":"2026-02-13T18:38:18Z","title_canon_sha256":"aac9a2b7a6cc1c7c0bb919def4b518c639cc06088b59ca415dd6b7dffbf9d1e0"},"schema_version":"1.0","source":{"id":"2602.13379","kind":"arxiv","version":2}},"canonical_sha256":"6d98240210bf679f6eb6490ee1f8518baa7528477756da1dfea5ce2ce081ebe9","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6d98240210bf679f6eb6490ee1f8518baa7528477756da1dfea5ce2ce081ebe9","first_computed_at":"2026-06-12T01:08:22.796038Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-12T01:08:22.796038Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"0YVaxsNXCstfEgEKuYCBM9kULrSsOm8d3/P8BOsI9cRLTDif1vaapZY3jMOYex7vSaTzAY4yWr1IeMV0CNuiCw==","signature_status":"signed_v1","signed_at":"2026-06-12T01:08:22.797107Z","signed_message":"canonical_sha256_bytes"},"source_id":"2602.13379","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ad1b0f8f088f32038e4b71eaa393f3c142b3d7343644daa2a384a9b5396e3430","sha256:40ff6123d60e14118d99f29307cb071564ad69b4720b108d2b8b83314c2d24af"],"state_sha256":"a85c7decf801234e7326766c11f5a7e7421f38155d2f69548f0eb459238480df"}